HR_Analytics

Human Resources Analytics: Why are our best employees leaving?

View the Project on GitHub

Using an open source data set from Kaggle containing employee features such as satisfaction rates, salary, and retention to determine what kinds of employees are leaving the company and why.

Data Set Features:

Satisfaction Level, Last Evaluation, Number of Projects, Average Monthly Hours, Time Spent at Company, Work Accident, Quit Company, Promotion in Last 5 Years, Department, Salary

The Big Picture: An Overview of Past and Present Employees

screen shot 2017-01-17 at 4 37 03 pm 2

The Big Picture: Coorelation

264c19f0-dcd2-11e6-977a-7c5c14350550

Taking a look at the correlation of key variables highlighted above (whether the employee has left or not and how long they stayed at the company) with other employee attributes reveals charecteristics of those that leave the company:

Coorelation Matrix of Present Employees:

screen shot 2017-01-17 at 4 40 46 pm 2

The key variable in this matrix is the amount of time spent at the company. An employee will quit, as thier satisfaction levels, evaluation, # of projects and hours work decrease. They will work less hours as they score less in evaluations and are given less projects.

Comparing the descriptive statistics for both groups we discover: